Autonomous Vehicle Simulation (MDAS.ai)

Sridhar LakshmananDepartment of Electrical & Computer Engineering

University of Michigan - Dearborn

Presentation for Physical Systems Replication Panel –NDIA Cyber-Enabled Emerging Technologies Symposium

Emerging Technologies

Key words Faculty Involved

PerceptionBig Data

Machine learningBayesian Inference

Sensor fusion

Sridhar LakshmananYi Lu Murphey

Paul Watta

Intelligent ControlAutonomous vehicles

UAVIndustrial robots

Stan BaekYu Zheng

Samir RawashdehMichael Putty

Vehicle Communications v2vv2iv2p

Paul RichardsonWeidong XiangChun-Hung Liu

StandardsSAE On-Road Automated Vehicle Systems (J3016) / Functional Safety (ISO 26262)

RVSWG 20 light and medium trucks standardSteve Underwood

Mark Zachos

CybersecurityFingerprinting ECU’s

IDSHafiz Malik

Di Ma

Power ElectronicsSolid state convertors

Electric drivesCharging

Kevin (Hua) Bai Maggie WangTaehyung KimWencong Su

Sensors & ChipsChip Design / SOCNano technologySolid state optics

Riadul IslamAlex Yi

Core Areas of Expertise

AU

TON

OM

YELEC

TRO

NIC

S

Sridhar LakshmananPh.D. Electrical & Computer Eng. (UMass-Amherst)

Associate Professor

University of Michigan–Dearborn

Office: SFC 212

313.593.5516 (O)

734.646.8920 (M)

lakshman@umich.edu

linkedin.com/in/slakshmanan

researchgate.net/profile/Sridhar_Lakshmanan

http://www.MDAS.ai

Autonomous Navigation: Army ATD

Driver Monitoring: NHTSA

Sensor Fusion: DARPA

Pedestrian Detection: Ford URP

Miniature Robots: Army SBIR

Lane Detection: Army

Design, Build & Test

Why simulate?

Bring data back

Requirements

Failure modes

MDAS.ai Timeline & Ecosystem

April ‘18: MI Robotics Day

Aug ‘18: UMD-MEDC Showcase

Dec ’18: v1.0 Shuttle (Straightaway)

May ’19: AutoSens-D

Fall ‘19: v2.0 Shuttle (Loop)

Drive-by-wire conversion

Power-assisted steer

Linear brake

Analog throttle

Localization

Sub-cm accuracy

GPS+

Deep learning: Nvidia GPU

Tracking

IntentionLocation

> Import 3-D models of the environment / build one in real-time

> Determine availability, accuracy of geo-location: GPS INS IMU

> Introduce a variety of dynamic 3-D actors into the environment: people, vehicles, animals, etc.

> Individually program the trajectory for each of these actors: location, path, speed, timing, etc.

> Deploy a library sensors: cameras, stereo heads, IR cameras, LIDARs, FMCW Radars, Ultrasound sensors

> Fine tune these sensors at component level: optics, electronics, illumination, etc.

> Model environmental conditions that affect sensing: weather, lighting, smoke, etc.

> Deploy a library of algorithms, including open-source (ROS): path, static /dynamic objects

> Specific a mobility mission: Leader-follower, point-to-point, path, speed, time, etc.

> Deploy a library of control algorithms, including open-source (ROS) ones, to meet mission objectives

> Quantify mobility: Speed vs. Trafficability, Time vs. Situational Awareness, etc.

> Select a vehicle and associated power-/drive-train: electric, hybrid, fuel, etc.

> Articulate motion: traction, brake, throttle, steering, teleop, etc.

Environment

Perception

Control

VehiclePerformance

High-Fidelity Modeling and Simulation of Complex Pedestrian and Traffic for Supervised Teleop Convoys High-fidelity simulation

Co

mp

ute

r M

od

el

Physics-based Computer Models of Sensors and

Comms – Camera, Radar, LIDAR, GPS, Li-Fi, RF

Physics-based Non-linear Vehicle Dynamics Model –Steering, Throttle, Braking,

Teleop

Perception-based Control with Special Emphasis on Pedestrian

Detection and Tracking, and Intention Estimation

Per

form

ance

Met

rics

Mobility ModelMulti-Disciplinary Project

Capabilities

Campus mobility model is Physics-based and not based on empirical data (see next sheet)

Special case of the Next-Generation NATO Reference Mobility Model (NG-NRMM)

Computer model is validated by real data from the physical shuttle MDAS.ai, and conversely, computer model is used to improve on-road performance of the vehicle

Model output is performance metrics such as – Mobility, Traversability, Repeatability, Reliability

Model used to:

Assess and compare autonomous systems in campus/urban environments

Compare autonomous systems to baseline human-driven systems

Benchmark progression of autonomous systems from Level-0 to Level-5

Assess performance of Perception Systems and Control Strategies

Physical System

Computer Model

Performance Metrics

High-Fidelity Simulation:System of Systems of Systems

> Import 3-D models of the environment / build one in real-time

> Determine availability, accuracy of geo-location: GPS INS IMU

> Introduce a variety of dynamic 3-D actors into the environment: people, vehicles, animals, etc.

> Individually program the trajectory for each of these actors: location, path, speed, timing, etc.

> Deploy a library sensors: cameras, stereo heads, IR cameras, LIDARs, FMCW Radars, Ultrasound sensors

> Fine tune these sensors at component level: optics, electronics, illumination, etc.

> Model environmental conditions that affect sensing: weather, lighting, smoke, etc.

> Deploy a library of algorithms, including open-source (ROS): path, static /dynamic objects

> Specific a mobility mission: Leader-follower, point-to-point, path, speed, time, etc.

> Deploy a library of control algorithms, including open-source (ROS) ones, to meet mission objectives

> Quantify mobility: Speed vs. Trafficability, Time vs. Situational Awareness, etc.

> Select a vehicle and associated power-/drive-train: electric, hybrid, fuel, etc.

> Articulate motion: traction, brake, throttle, steering, teleop, etc.

Environment

Perception

Control

VehiclePerformance

High-Fidelity Modeling and Simulation of Complex Pedestrian and Traffic for Supervised Teleop Convoys